Scientists have identified molecules in blood and urine that can reveal how much energy a person consumes from ultra-processed foods, offering a groundbreaking tool for nutrition research. The study, led by Erikka Loftfield from the US National Cancer Institute and published in PLOS Medicine, marks the first time that biological markers have been linked directly to ultra-processed food intake. This breakthrough could provide a more accurate and objective method of assessing diet quality, bypassing the unreliable nature of traditional self-reported dietary data.
Ultra-processed foods—such as packaged snacks, sugary cereals, soft drinks, and frozen meals—comprise nearly 60 percent of the American diet and are associated with various chronic diseases. However, quantifying their intake has been a major challenge in nutrition science. Standard dietary studies depend on participants recalling what they ate, often leading to misreporting or omissions.
Loftfield and her team analyzed data from over 1,000 older adults in the AARP cohort, including more than 700 participants who provided both dietary recall data and blood and urine samples. The researchers discovered that hundreds of metabolites—substances produced during digestion and metabolic processes—were linked to energy derived from ultra-processed foods. From this, they developed a predictive score using 28 blood markers and up to 33 urine markers.
These markers proved reliable in identifying overall dietary patterns rather than individual food items. In particular, two amino acids and a carbohydrate were repeatedly identified as key indicators. One of these markers was also associated with type 2 diabetes, suggesting a possible link between high consumption of ultra-processed foods and metabolic disorders.
To validate their findings, the team applied the scoring system to a tightly controlled 2019 NIH study in which 20 adults lived on-site and alternated between ultra-processed and unprocessed diets. The metabolite scores successfully distinguished between high and low consumption periods, confirming the tool’s accuracy at an individual level.
Experts say this is a major advance in nutritional science. Dr. Dariush Mozaffarian of Tufts University emphasized the potential for this method to explore how different types of ultra-processed foods and their components affect health outcomes. Loftfield hopes to use the tool to examine links between ultra-processed food intake and chronic diseases like cancer. As scientific and public interest in diet-related health grows, funding these studies remains a critical next step.
READ MORE: